py_ciu_image : A Python Library for Explaining Image Classification with Contextual Importance and Utility

Kary Främling; Ioan Vlad Apopei; Gustav Grund Pihlgren; Avleen Malhi

doi:10.1007/978-3-031-70074-3_10

py_ciu_image : A Python Library for Explaining Image Classification with Contextual Importance and Utility

Kary Främling^*, Ioan Vlad Apopei, Gustav Grund Pihlgren, Avleen Malhi

^*Corresponding author for this work

Department of Industrial Engineering and Management

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

Abstract

Contextual Importance and Utility (CIU) is a model-agnostic method for explaining outcomes of AI systems. CIU has succeeded in producing meaningful explanations where state-of-the-art methods fail, e.g. for detecting bleeding in gastroenterological images. This paper presents a Python implementation of CIU for explaining image classifications.

Original language	English
Title of host publication	Explainable and Transparent AI and Multi-Agent Systems - 6th International Workshop, EXTRAAMAS 2024, Revised Selected Papers
Editors	Davide Calvaresi, Amro Najjar, Andrea Omicini, Rachele Carli, Giovanni Ciatto, Reyhan Aydogan, Joris Hulstijn, Kary Främling
Place of Publication	Cham
Publisher	Springer
Pages	184-188
Number of pages	5
ISBN (Electronic)	978-3-031-70074-3
ISBN (Print)	978-3-031-70073-6
DOIs	https://doi.org/10.1007/978-3-031-70074-3_10
Publication status	Published - 2024
MoE publication type	A4 Conference publication
Event	International Workshop on Explainable, Transparent Autonomous Agents and Multi-Agent Systems - Auckland, New Zealand Duration: 6 May 2024 → 10 May 2024 Conference number: 6

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Publisher	Springer
Volume	14847 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Workshop

Workshop	International Workshop on Explainable, Transparent Autonomous Agents and Multi-Agent Systems
Abbreviated title	EXTRAAMAS
Country/Territory	New Zealand
City	Auckland
Period	06/05/2024 → 10/05/2024

Keywords

Contextual Importance and Utility
Deep Neural Network
Explainable Artificial Intelligence
Image Classification

Access to Document

10.1007/978-3-031-70074-3_10

Cite this

Främling, K., Apopei, I. V., Pihlgren, G. G., & Malhi, A. (2024). py_ciu_image : A Python Library for Explaining Image Classification with Contextual Importance and Utility. In D. Calvaresi, A. Najjar, A. Omicini, R. Carli, G. Ciatto, R. Aydogan, J. Hulstijn, & K. Främling (Eds.), Explainable and Transparent AI and Multi-Agent Systems - 6th International Workshop, EXTRAAMAS 2024, Revised Selected Papers (pp. 184-188). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 14847 LNAI). Springer. https://doi.org/10.1007/978-3-031-70074-3_10

Främling, Kary ; Apopei, Ioan Vlad ; Pihlgren, Gustav Grund et al. / py_ciu_image : A Python Library for Explaining Image Classification with Contextual Importance and Utility. Explainable and Transparent AI and Multi-Agent Systems - 6th International Workshop, EXTRAAMAS 2024, Revised Selected Papers. editor / Davide Calvaresi ; Amro Najjar ; Andrea Omicini ; Rachele Carli ; Giovanni Ciatto ; Reyhan Aydogan ; Joris Hulstijn ; Kary Främling. Cham : Springer, 2024. pp. 184-188 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{6925cb62b36a41b094dffa07d606f946,

title = "py_ciu_image : A Python Library for Explaining Image Classification with Contextual Importance and Utility",

abstract = "Contextual Importance and Utility (CIU) is a model-agnostic method for explaining outcomes of AI systems. CIU has succeeded in producing meaningful explanations where state-of-the-art methods fail, e.g. for detecting bleeding in gastroenterological images. This paper presents a Python implementation of CIU for explaining image classifications.",

keywords = "Contextual Importance and Utility, Deep Neural Network, Explainable Artificial Intelligence, Image Classification",

author = "Kary Fr{\"a}mling and Apopei, {Ioan Vlad} and Pihlgren, {Gustav Grund} and Avleen Malhi",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.; International Workshop on Explainable, Transparent Autonomous Agents and Multi-Agent Systems, EXTRAAMAS ; Conference date: 06-05-2024 Through 10-05-2024",

year = "2024",

doi = "10.1007/978-3-031-70074-3_10",

language = "English",

isbn = "978-3-031-70073-6",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer",

pages = "184--188",

editor = "Davide Calvaresi and Amro Najjar and Andrea Omicini and Rachele Carli and Giovanni Ciatto and Reyhan Aydogan and Joris Hulstijn and Kary Fr{\"a}mling",

booktitle = "Explainable and Transparent AI and Multi-Agent Systems - 6th International Workshop, EXTRAAMAS 2024, Revised Selected Papers",

address = "Germany",

}

Främling, K, Apopei, IV, Pihlgren, GG & Malhi, A 2024, py_ciu_image : A Python Library for Explaining Image Classification with Contextual Importance and Utility. in D Calvaresi, A Najjar, A Omicini, R Carli, G Ciatto, R Aydogan, J Hulstijn & K Främling (eds), Explainable and Transparent AI and Multi-Agent Systems - 6th International Workshop, EXTRAAMAS 2024, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 14847 LNAI, Springer, Cham, pp. 184-188, International Workshop on Explainable, Transparent Autonomous Agents and Multi-Agent Systems, Auckland, New Zealand, 06/05/2024. https://doi.org/10.1007/978-3-031-70074-3_10

py_ciu_image : A Python Library for Explaining Image Classification with Contextual Importance and Utility. / Främling, Kary; Apopei, Ioan Vlad; Pihlgren, Gustav Grund et al.
Explainable and Transparent AI and Multi-Agent Systems - 6th International Workshop, EXTRAAMAS 2024, Revised Selected Papers. ed. / Davide Calvaresi; Amro Najjar; Andrea Omicini; Rachele Carli; Giovanni Ciatto; Reyhan Aydogan; Joris Hulstijn; Kary Främling. Cham: Springer, 2024. p. 184-188 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 14847 LNAI).

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - py_ciu_image : A Python Library for Explaining Image Classification with Contextual Importance and Utility

AU - Främling, Kary

AU - Apopei, Ioan Vlad

AU - Pihlgren, Gustav Grund

AU - Malhi, Avleen

N1 - Conference code: 6

PY - 2024

Y1 - 2024

N2 - Contextual Importance and Utility (CIU) is a model-agnostic method for explaining outcomes of AI systems. CIU has succeeded in producing meaningful explanations where state-of-the-art methods fail, e.g. for detecting bleeding in gastroenterological images. This paper presents a Python implementation of CIU for explaining image classifications.

AB - Contextual Importance and Utility (CIU) is a model-agnostic method for explaining outcomes of AI systems. CIU has succeeded in producing meaningful explanations where state-of-the-art methods fail, e.g. for detecting bleeding in gastroenterological images. This paper presents a Python implementation of CIU for explaining image classifications.

KW - Contextual Importance and Utility

KW - Deep Neural Network

KW - Explainable Artificial Intelligence

KW - Image Classification

UR - http://www.scopus.com/inward/record.url?scp=85206108952&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-70074-3_10

DO - 10.1007/978-3-031-70074-3_10

M3 - Conference article in proceedings

AN - SCOPUS:85206108952

SN - 978-3-031-70073-6

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 184

EP - 188

BT - Explainable and Transparent AI and Multi-Agent Systems - 6th International Workshop, EXTRAAMAS 2024, Revised Selected Papers

A2 - Calvaresi, Davide

A2 - Najjar, Amro

A2 - Omicini, Andrea

A2 - Carli, Rachele

A2 - Ciatto, Giovanni

A2 - Aydogan, Reyhan

A2 - Hulstijn, Joris

A2 - Främling, Kary

PB - Springer

CY - Cham

T2 - International Workshop on Explainable, Transparent Autonomous Agents and Multi-Agent Systems

Y2 - 6 May 2024 through 10 May 2024

ER -

Främling K, Apopei IV, Pihlgren GG, Malhi A. py_ciu_image : A Python Library for Explaining Image Classification with Contextual Importance and Utility. In Calvaresi D, Najjar A, Omicini A, Carli R, Ciatto G, Aydogan R, Hulstijn J, Främling K, editors, Explainable and Transparent AI and Multi-Agent Systems - 6th International Workshop, EXTRAAMAS 2024, Revised Selected Papers. Cham: Springer. 2024. p. 184-188. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-031-70074-3_10